{
 "measured_at": "2026-07-11",
 "judge_model": "claude-sonnet-5 (Claude Code Agent tool, isolated sub-agents, batch of 6 items, 4 shells + 2 domain-native per batch)",
 "judge_input": "verbatim '## Wording-Pattern Advisory' section from deep-research/agents/socratic_mentor_agent.md (pre-#503 variant = commit c9728d6^; post-#503 variant = current), plus the six batch items; no other context",
 "decision_rule": "advisory fires / stays silent per item, judged independently",
 "thresholds_inherited_from": "evals/gold/rq_framing_patterns/manifest.yaml (FNR < 0.30, FPR < 0.20)",
 "runs": {
  "pre_503": {
   "nat-065": true,
   "nat-077": false,
   "dn-006": false,
   "dn-005": false,
   "nat-028": true,
   "nat-018": true,
   "dn-007": false,
   "nat-049": false,
   "ti-007": true,
   "nat-044": false,
   "dn-016": false,
   "nat-014": true,
   "dn-008": false,
   "nat-012": true,
   "el-003": true,
   "nat-045": true,
   "el-013": false,
   "dn-009": false,
   "nat-048": true,
   "ti-008": false,
   "dn-002": false,
   "el-007": true,
   "el-017": false,
   "dn-018": false,
   "dn-013": false,
   "ti-013": false,
   "nat-001": true,
   "nat-021": true,
   "el-016": true,
   "dn-011": false,
   "dn-001": false,
   "el-001": true,
   "nat-061": false,
   "dn-019": false,
   "ti-010": false,
   "ti-001": false,
   "ti-012": false,
   "nat-059": false,
   "nat-004": true,
   "dn-012": false,
   "ti-004": true,
   "ti-002": true,
   "el-010": false,
   "dn-014": false,
   "dn-003": false,
   "nat-053": true,
   "nat-037": true,
   "el-005": true
  },
  "post_503_rep1": {
   "nat-065": true,
   "nat-077": true,
   "dn-006": false,
   "dn-005": false,
   "nat-028": true,
   "nat-018": true,
   "dn-007": false,
   "nat-049": false,
   "ti-007": true,
   "nat-044": false,
   "dn-016": false,
   "nat-014": true,
   "dn-008": false,
   "nat-012": true,
   "el-003": true,
   "nat-045": true,
   "el-013": false,
   "dn-009": false,
   "nat-048": true,
   "ti-008": false,
   "dn-002": false,
   "el-007": true,
   "el-017": false,
   "dn-018": false,
   "dn-013": false,
   "ti-013": false,
   "nat-001": true,
   "nat-021": true,
   "el-016": true,
   "dn-011": false,
   "dn-001": false,
   "el-001": true,
   "nat-061": false,
   "dn-019": false,
   "ti-010": false,
   "ti-001": true,
   "ti-012": false,
   "nat-059": false,
   "nat-004": true,
   "dn-012": false,
   "ti-004": false,
   "ti-002": false,
   "el-010": false,
   "dn-014": false,
   "dn-003": false,
   "nat-053": true,
   "nat-037": true,
   "el-005": true
  },
  "post_503_rep2": {
   "nat-065": true,
   "nat-077": true,
   "dn-006": false,
   "dn-005": false,
   "nat-028": true,
   "nat-018": true,
   "dn-007": false,
   "nat-049": true,
   "ti-007": true,
   "nat-044": false,
   "dn-016": false,
   "nat-014": true,
   "dn-008": false,
   "nat-012": true,
   "el-003": true,
   "nat-045": true,
   "el-013": false,
   "dn-009": false,
   "nat-048": true,
   "ti-008": false,
   "dn-002": false,
   "el-007": true,
   "el-017": false,
   "dn-018": false,
   "dn-013": false,
   "ti-013": false,
   "nat-001": true,
   "nat-021": true,
   "el-016": true,
   "dn-011": false,
   "dn-001": false,
   "el-001": true,
   "nat-061": false,
   "dn-019": false,
   "ti-010": false,
   "ti-001": true,
   "ti-012": false,
   "nat-059": false,
   "nat-004": true,
   "dn-012": false,
   "ti-004": false,
   "ti-002": false,
   "el-010": false,
   "dn-014": false,
   "dn-003": false,
   "nat-053": true,
   "nat-037": true,
   "el-005": true
  }
 },
 "summary": {
  "pre_503": {
   "miss_rate": 0.375,
   "miss_family_variant": 0.261,
   "miss_off_list": 0.667,
   "false_fire": 0.0
  },
  "post_503_rep1": {
   "miss_rate": 0.375,
   "miss_family_variant": 0.217,
   "miss_off_list": 0.778,
   "false_fire": 0.0
  },
  "post_503_rep2": {
   "miss_rate": 0.344,
   "miss_family_variant": 0.174,
   "miss_off_list": 0.778,
   "false_fire": 0.0
  },
  "stable_misses_post": [
   "el-010",
   "el-013",
   "el-017",
   "nat-044",
   "nat-061",
   "ti-002",
   "ti-004",
   "ti-008",
   "ti-010",
   "ti-012",
   "ti-013"
  ],
  "unstable_items_post": [
   "nat-049"
  ]
 },
 "verdict": "MISS RATE HIGH (overall FNR 0.34-0.38 >= 0.30 line; off-list tier 0.78 stable across replicates; FPR 0.00). Per issue #501, this set becomes the acceptance test for any future change to the advisory."
}